Unintended benefits from the next phase of financial sector reforms

Transcription

1 Deutsche Bank Markets Research Asia China Banking / Finance Industry Chinese Stay Positive Date Industry Update Unintended benefits from the next phase of financial sector reforms Asset quality concerns overdone; positive stance maintained We believe asset quality concerns for the H-share listed banks, especially the big four banks, are overdone given the recovering economy. Our cash flow analysis does not suggest under-reporting of NPLs, an outcome further confirmed by interest receivables accounting for only 46bps of the banks' assets, comparable to the Hong Kong banks. In addition to the falling perceived asset quality risks, we expect the next phase of financial sector reform to drive market risks down, unintentionally benefiting the bank share prices, although selected joint stock banks like MSB might be more affected by tighter regulation on inter-bank exposure and deposit rate deregulation. Unintended benefits from the next phase of financial sector reforms We believe that the new reform policies announced by regulators, such as the issuance of preferred shares, asset securitization and the formation of free trade zone in Shanghai, are positive developments for the H-share listed banks. In addition to the obvious benefits of newer businesses and reduced capital pressure on banks, we see the emergence of new market-driven benchmarks (e.g. lower-than-expected yields of preferred shares issued by the big four banks) driving the market risks lower and unintentionally benefiting the share prices of the banks. Our preferred shares are extremely long-dated and noncumulative tier 1 debt instruments that are subject to investment risks such as skipping of payments and forced conversion if the earnings and core tier 1 (CT1) ratio fall short of target levels. Cash flow analysis demystifies asset quality concerns of H-share banks In this report, we have extensively analyzed the asset quality of Chinese banks and, assuming a stable economy, expect risks to be low despite the growth of the shadow banking system. Most importantly, our cash flow analysis does not suggest under-reporting of NPLs, as the increase in PBT of the H-share listed banks from 2007 to 2012 is matched by the net increase in cash. The analysis is further confirmed by accrued interest income accounting for 46bps of the assets of these banks (versus 74bps for India, 57bps for HK), making us confident about the asset quality of listed banks. Stay positive despite tighter regulatory atmosphere; we prefer big four banks Under pressure to sustain profitability despite weaker deposit franchises, jointstock banks have been serving as intermediaries to the liquid banks and nonbank financial system, leading to a higher issuance of WMPs and increased financing for the shadow banking system through inter-bank transactions. Our analysis shows that if the CBRC were to raise the risk weighting of the financial assets sold and held under resale arrangements that were collateralized against bills to 75%, and that of trust beneficial rights to 100% to slow the flow, the CT1 for the H-share listed banks would be lowered by 23bps to 9.44%, with the bigger negative impact felt by joint-stock banks like MSB (- 1.32%) to 6.54% and CRCB (-1.42%) to 9.5%. Hence, we maintain the big four banks as our top picks, with our estimates showing the sector trading on 5.8x 2013E P/E and 1.1x 2013E P/B. We identify a slowing Chinese economy as the key risk to the sector s performance. Tracy Yu Sukrit Khatri Research Analyst Research Associate (+852) (+852) tracy.yu@db.com sukrit.khatri@db.com Top picks ICBC (1398.HK),HKD5.54 China Construction Bank (0939.HK),HKD6.14 Agri. Bank of China (1288.HK),HKD3.71 Bank of China (3988.HK),HKD3.60 Source: Deutsche Bank Companies Featured ICBC (1398.HK),HKD5.54 China Construction Bank (0939.HK),HKD6.14 Agri. Bank of China (1288.HK),HKD3.71 Bank of China (3988.HK),HKD3.60 Bank of Communications (3328.HK),HKD5.89 China Merchants Bank-H (3968.HK),HKD14.52 China CITIC Bank (0998.HK),HKD4.27 China Minsheng Bank (1988.HK),HKD9.67 Chongqing Rural Bank (3618.HK),HKD3.94 China Everbright Bank ( SS),CNY2.96 Source: Deutsche Bank Buy Buy Buy Buy Buy Buy Buy Buy Buy Buy Buy Hold Buy Hold Deutsche Bank does and seeks to do business with companies covered in its research reports. Thus, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision. DISCLOSURES AND ANALYST CERTIFICATIONS ARE LOCATED IN APPENDIX 1. MICA(P) 054/04/2013.

2 Chinese Stay Positive Table Of Contents Investment summary... 3 Stay positive; the next phase of financial sector reforms to drive market risks down... 3 Asset quality Why things are better than they look... 5 Chinese banking system is highly concentrated, with the three policy banks being non-profit seeking... 5 Reported NPLs, cash flow and interest accrual analysis on banks all suggest asset quality concerns are largely overdone... 7 Shadow banking system reasons, trends and risks Shorter loan duration, refinancing LGFV loans and role of joint-stock banks Trust companies are playing a key role in growth Relationship between banks and shadow banking system Joint-stock banks are more vulnerable to liquidity and credit risks of the shadow banking system Our proprietary research trips in China suggest risks are high; but risk management seems effective amid stable economy Shadow banking is transforming the risk profile of the banking system CBRC could effectively slow the flow Valuation and risks Valuation of Chinese banks Key risks for Chinese banks Appendix A Major regulations and policies of the trust business Page 2

3 Chinese Stay Positive Investment summary Stay positive; the next phase of financial sector reforms to drive market risks down In this report we highlight that the new reforms announced by Chinese regulators recently, including the issuance of preferred shares, asset securitization and the formation of free trade zone in Shanghai, should be positive to the Chinese banking sector. As such, we expect the share price performances of the H-share listed banks to be supported by the announcements of policy details over the course of the next few months. These reforms are expected to further deregulate the financial sector, the RMB exchange rate and deposit rates. In addition to the obvious benefits of creating newer businesses and reduced pressure on the issuance of common equities, we believe one of the most positive factors is the emergence of market-based prices to improve credit allocation and reduce distortions. In addition, we believe that some of the prices might become powerful benchmarks to evaluate the risks taken by the Chinese banks and indicate whether the current share prices are under- or overvalued. In our view, the market yields for preferred shares issued by the big four banks are the most relevant benchmarks, as these prices should indicate long-term risks that institutional investors are willing to take in terms of skipping of payments and forced conversion into common equity when earnings and the target core tier 1 ratio fall short of a target level. Our preferred shares are extremely long-dated tier 1 debt that is non-accumulative and does not carry a voting right. Given China s high savings rate and strong demand for fixed income instruments and the relatively low yields cleared for the dim sum bond market in Hong Kong, we see a high possibility that the yields of the preferred shares for the big four banks might be cleared at a level that is not significantly higher than their dividend yields. In this case, we believe the H-share listed banks might be re-rated on falling perceived asset quality risks, given their low valuation of 5.8x 2013E P/E and 1.1x 2013E P/B, and given their ROAE of 19.1%. Our earnings forecasts for FY2013 already incorporate the negative impact of the deposit pricing band widening to 1.2x (current: 1.1x) of the benchmark deposit rates. In this report, we provide detailed reasoning on why market concerns on the asset quality of the H-share listed banks seem to be overdone if China s economy continues to be stable. A summary of our key points can be found below: Our cash flow analysis suggests that NPLs are not under-reported by the H-share listed banks, and this is supported by the fact that accrued interest income accounts for only 46bps of assets, slightly better than that for the Hong Kong banks. As a result, we believe concerns about the asset quality of the H-share listed banks have been largely overdone. Tighter bank regulations since 2009 has fueled strong growth in shadow banking, which we estimate to be at Rmb12.5trn as of June 2013 (or 21.9% of 2013E China s GDP), a twofold rise in the last 2.5 Page 3

4 Chinese Stay Positive years. In particular, restrictions against bank lending to LGFVs and property developers have led to increased demand for credit from nonbanking institutions. Property, local government and small and micro enterprise exposures likely account for the bulk of shadow banking exposures. Our estimated size of the shadow banking system at Rmb12.5trn is smaller than forecasts by other commentators, as we primarily include credit granted by the non-bank financial institutions, which are riskier, and exclude issuance of corporate bonds and short-term corporate financing provided by banks through bank acceptance. Our proprietary research trips in China, where we met government officials and over 30 non-bank credit operators in Beijing, Shanghai, Hangzhou, Xi an, Shenyang, Hohhot, and Ordos, suggested that although the magnitude of default risk is higher for the less regulated non-bank financial system, the management of these risks seems entirely market-driven and generally effective. The shadow banking system seems to be funded primarily through (1) issuance of wealth management products by banks, reaching Rmb9.1trn as of June 2013 (one-third is invested in trust and other loans); (2) inter-bank financing, as indicated by sharply rising financial assets held by H-share banks under resale arrangements, reaching Rmb5.2trn as of June 2013; and (3) some small proprietary investments made by the banks, totaling Rmb140.4bn by the H-share listed banks as of June The largest segment of the shadow banking system is the trust sector, which has seen AUM grow to Rmb9.5trn as of June 2013 (of which 44% were trust loans), doubling in 18 months. Trusts are regulated by the CBRC, with key players backed by large financial institutions as their major shareholders. The systemic concerns raised by the growth of shadow banking arise from the reliance of shadow banks on funding from the banking system, as this high connectivity means credit or liquidity risks in shadow banking could evolve into counterparty risks for the banks. Amongst the listed banks, the largest commercial banks (ICBC, CCB, ABC, BOC and BoCom, which account for 45% of banking system assets) have proportionately the lowest exposure to shadow banking. The three policy banks, which account for about 8% of assets, have pure sovereign risk. The risks appear to be greatest for the joint-stock banks (18% of assets), which are the main intermediaries between the banking system and the non-banks. They have disproportionately large exposures to WMPs and to inter-bank assets collateralized against bills and trust beneficial rights. Unsurprisingly, this group of banks was at greatest risk during the Shibor shock in June as wholesale funding dried up. We believe the CBRC could effectively slow the flow to the shadow banking system by increasing the risk weightings on resale agreements. Our analysis shows that if the weighting on resale agreements collateralized against bills was raised to 75% and that of trust beneficial rights to 100%, the CT1 of the H-share listed banks would be lowered by 23bps to 9.44%, with the biggest negative impact felt by joint-stock banks like MSB (-1.32% to 6.54%) and CRCB (-1.42% to 9.5%). Page 4

5 Chinese Stay Positive Asset quality Why things are better than they look Chinese banking system is highly concentrated, with the three policy banks being non-profit seeking In the absence of a strong and liquid capital market, China s financial system is heavily dependent on the banking system, which consists of a highly concentrated network of banks, finance companies and other banking institutions. China s regulated financial system, which has Rmb133.4trn in assets and Rmb1.5trn in profits, is dominated by the big five commercial banks (ICBC, CCB, ABC, BOC, and BoCom partly state-owned, partly privatized), which enjoy over 50% of the system s net profit, while employing 45% of the total system assets and 46% of listed equity (as of December 2012). As Figure 1 shows, the next layer of the regulated financial system is made up of the national joint-stock commercial banks, 12 in total, which are neither wholly state-owned nor wholly private, and sometimes controlled by SOEs, with the help of minority foreign bank shareholders. These make up 18% of total assets, 15% of total equity and 17% of net profit. In addition, there exist smaller financial institutions in the form of city commercial banks (144) and rural financial institutions (2,411), which are dispersed throughout the country with influence restricted to their respective locations. Figure 1: An overview of the regulated banking system in China (as of and for the year ending 31 December 2012) Large Commercial National Joint Stock Commercial No. of legal institutions Total assets (RMB bn) Shareholders' equity (RMB bn) Net profit (RMB bn) Total amount Mkt shr Total amount Mkt shr Total amount Mkt shr NPL ratio Loss ratio 5 60,040 45% 3,952 46% % 1.0% 0.1% 12 23,527 18% 1,314 15% % 0.7% 0.1% Policy 3 11,217 8% 553 6% 74 5% City Commercial Rural Financial Institutions Foreign Banking Institutions Other Banking Institutions ,347 9% 808 9% 137 9% 0.8% 0.1% 2,411 15,512 12% % % 1.8% 0.1% 42 2,380 2% 256 3% 16 1% 0.5% 0.1% 1,130 8,599 6% 792 9% 116 8% Total 3, , % 8, % 1, % 1.0% 0.1% Source: Deutsche Bank There are also three policy banks (China Development Bank, Export Import Bank and Agricultural Development Bank) that are wholly government-owned, non-profit and primarily engaged in lending for infrastructure, trade and agriculture respectively. As a result, the LDR ratios of these banks are not comparable with the commercial banks in China, and the asset quality of the policy banks are not constrained by their ability to take deposits. Page 5

6 Chinese Stay Positive Figure 2: Policy banks: Loans and deposits outstanding vs. Gross loans as a % of GDP Loans Outstanding (LHS) Deposits Outstanding (LHS) Rmb tn Loans as % of GDP (RHS) % 20% % % 17.2% 18% % 14.5% % % % % % 0.0 8% Source: Deutsche Bank,CEIC, company data Figure 3: Gross loans for policy banks in China ( ) Rmb Bn China Development Bank 3,708 4,510 5,526 6,418 Agri. Development Bank of China 1,451 1,671 1,876 2,185 Export Import Bank ,183 Total 5,752 6,888 8,316 9,786 Source: Deutsche Bank, company data As of December 2012, the three policy banks gave out Rmb9.8trn in loans, but made only Rmb74bn in profits. The bulk of these loans are LGFV loans, with losses to be absorbed by the state. Figures 4 to 6 provide loan breakdowns for the three policy banks as of FY12. Figure 4: Loan breakdown of China Development Bank (FY12) China Development Bank (FY12 Loans: Rmb 6,418bn) Figure 5: Loan breakdown of Export Import Bank (FY12) Exim Bank (FY12 Loans: Rmb 1,183bn) Figure 6: Loan breakdown of Agri. Development Bank of China (FY12) Agri Development Bank of China (FY12 Loans: Rmb 2,185bn) Petroleum, Petrochemical and chemical industry, 529, 8% Others, 2,061, 32% Transportation, 1,942, 30% Public utilities & environment protection, 1,886, 30% Others, 376, 32% Import credit, 270, 23% Export seller's credit, 378, 32% Export buyer credit, 159, 13% Others, 1,288.8, 59% Grain and edible oil, 896.2, 41% Source: Deutsche Bank, company data, CEIC Source: Deutsche Bank, company data, CEIC Source: Deutsche Bank, company data, CEIC Operating in the shadows of the banking system in China are a host of other financial institutions which make up the non-traditional financial system, or the shadow banking system, and comprise both regulated (asset management companies, trust companies, money leasing and brokerage firms) and less regulated companies (pawn shops, peer-to-peer lenders, credit guarantee companies and leasing companies). The shadow banking system has grown rapidly in size and influence over the past few years, with total credit issued by these companies rising over two-fold in the last 2.5 years to reach Rmb12.5trn as of June 2013, according to Deutsche Bank estimates. Our estimate for the size of the shadow banking system is smaller than forecasts provided by other commentators, as we primarily include credit granted by the non-bank financial institutions, which are riskier, and exclude issuance of corporate bonds and short-term corporate financing provided by banks through bank acceptance. Page 6

7 Chinese Stay Positive Reported NPLs, cash flow and interest accrual analysis on banks all suggest asset quality concerns are largely overdone Despite the strong asset quality displayed by the H-share banks in the 1H13 results, with NPL growth slowing to 3.3% qoq in 2Q13 (versus 4.97% qoq in 1Q13), market concerns continue to persist on the credibility of the NPL ratio of 0.97% for the sector. To allay those concerns, we have undertaken a cash flow analysis, which does not suggest under-reporting of NPLs, as the increase in the pre-tax profit of the H-share listed banks from 2007 to 2012 was matched by the net increase in cash. This suggests that loans performed normally during that time, with asset quality concerns and doubts surrounding the credibility of the NPL ratio largely overdone. If Chinese banks under-report a notable amount of NPLs by classifying them as normal loans, we expect the sector s reported net profit to be higher than its cash profits, as interest income is accrued instead of paid in cash. Figure 7: Chinese banks: Ratio of net increase in cash adjusted for dividends to profit before tax Rmb bn Total Profit before income tax ,058 1,201 4,640 Net cash flow from operation 798 1, ,463 2,067 7,792 Net cash flow from investing , ,055-4,451 Net cash flow from financing Net increase in cash & cash equivalent -2 1, ,641 1,032 3,828 Dividends ,046 Net increase in cash adjusted for dividends 101 1, ,839 1,318 4,908 Net increase in cash/pbt 21% 340% -94% 62% 174% 110% 106% Source: Deutsche Bank, company data Note: Companies covered include ICBC, CCB, ABC, BOC, BOCOM, CMB, MSB, CNCB, CRCB and CEB Analyzing the ratio by bank, we see that the ratio is close to unity for the big four banks (ranging from 86% to 120%), while the smaller shareholding banks display more volatility. Figure 8: Chinese banks: Ratio of net increase in cash to PBT (from 2007 to 2012) 250% 199% 200% 184% 172% Figure 9: Volatility displayed by the ratio of net increase in cash and PBT BoCom CMB MSB CNCB CRCB CEB 1000% 800% 150% 100% 50% 0% 120% 103% 102% 86% 48% 46% 6% 106% 600% 400% 200% 0% -200% -400% Total Source: Deutsche Bank, company data Source: Deutsche Bank, company data Page 7

8 Chinese Stay Positive Our view is further supported by our analysis of accrued interest income (or interest receivables), which only accounted for 46bps of the assets of the H- share banks, with a range from 30bps (Minsheng Bank) to 52bps (CRCB). On a regional basis, Chinese banks performed better than banks in Hong Kong (57bps) and India (74bps), suggesting that there is nothing untoward about the asset quality of the H-share banks. Figure 10: Chinese banks: Interest receivables as % of assets (FY12) 0.55% 0.50% 0.45% 0.40% 0.35% 0.30% 0.25% 0.52% 0.50% 0.49% 0.46% 0.45%0.44%0.44%0.43% 0.42% 0.38% 0.30% 0.46% Figure 11: Asia: Interest receivables as % of assets (FY12) 0.80% 0.74% 0.70% 0.60% 0.57% 0.52% 0.50% 0.46% 0.39% 0.40% 0.29% 0.33% 0.30% 0.24% 0.20% 0.17% 0.18% 0.10% 0.00% TH TW SG AU ID KR CH PH HK IN Source: Deutsche Bank, company data Source: Deutsche Bank, company data Shadow banking system growing in size and influence As mentioned above, the shadow banking system has grown rapidly in size and influence over the past two years, to reach a total of Rmb12.5trn in June, a twofold rise in the last 2.5 years. As Figures 12 and 13 show, trust loans make up the majority of the shadow banking system, comprising 34% of total loans, followed by unregulated private lending (33%). Figure 12: Growth in credit issued by the shadow banking system in China RMB bn 14,000 12,000 10,000 8,000 6,000 4,000 2,000 0 Unregulated portion 5, ,573 2,400 7, ,735 3,380 10,390 1, ,999 12,524 1, ,215 3,718 4, ,300 1,460 1, FY10 FY11 FY12 1H13 Source: Deutsche Bank, CEIC, PBOC Financial leasing company Small-loan company Trust loan Private lending Credit guarantees Pawn loans Figure 13: Composition of credit issued by shadow banking system 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 12.0% 11.9% 14.9% 15.2% 3.4% 5.0% 27.0% 22.3% 41.3% 43.4% 5.7% 5.6% 28.9% 35.8% 33.7% 32.7% 15.7% 16.7% 14.0% 12.4% 0.6% 0.7% 0.7% 0.5% FY10 FY11 FY12 1H13 Source: Deutsche Bank, CEIC, PBOC Financial leasing company Small-loan company Trust loan Private lending Credit guarantees Pawn loans In order to account for the additional credit provided to the financial system by the non-bank institutions, PBOC reports a broader measure of credit available to the economy, total social financing (TSF), which posted a CAGR of 23% over the past decade ( ). While the banking component of TSF registered a CAGR of 17% during this period, non-bank financing witnessed a CAGR of 54%, and constituted 44% of the total new social financing in 1H13 (up from 5% in 2002). Page 8

9 Chinese Stay Positive Figure 14: Absolute level of total social financing (2002-1H13) Non-bank financing RMB trn 16.0 CAGR (FY02-FY12) 14.0 Total social financing: 23% Bank financing: 17% 12.0 Non-bank financing: 54% Bank financing FY02 FY03 FY04 FY05 FY06 FY07 FY08 FY09 FY10 FY11 FY12 1H13 Source: Deutsche Bank, CEIC Figure 15: Breakdown of total social financing into bank and non-bank financing % Non-bank financing Bank financing FY02 FY03 FY04 FY05 FY06 FY07 FY08 FY09 FY10 FY11 FY12 1H13 Source: Deutsche Bank, CEIC Page 9

10 Chinese Stay Positive Shadow banking system reasons, trends and risks Shorter loan duration, refinancing LGFV loans and role of joint-stock banks The unbridled growth in the shadow banking system has been primarily brought about by the shortening loan duration over the past two years. Over the past two years, the average loan duration of the banks has shortened from 2.43 years in 2010 to 2.22 years in 2012, reflecting a move towards shorterdated bank loans. The trend is confirmed by the split of new loans, with only 53% of new loans being medium-to-long term in 1H13, as against 86% in (Figure 17) Figure 16: The average loan duration of the H-share listed banks Years ICBC CCB ABC BOC BoCom CMB CNCB MSB CRCB Average Source: Deutsche Bank, company data In the absence of changes in the underlying economy, the shortening corporate loan duration has forced corporates to rely on non-bank modes of financing (bonds, trust and entrusted loans, microfinance issued by small money lenders) to bridge the funding mismatch gap. As a result, alternative financing has steadily gained in importance, and as of June it made up 44% of the total new social financing entering the financing system (up from 5% in 2002). Page 10

11 Chinese Stay Positive Figure 17: Mix change in the tenure of new bank loans in China New short-term loan New medium-to-long term loan 100% 90% 80% 38% 45% 53% 70% 59% 73% 60% 86% 50% Figure 18: Maturity profile of LGFV loans RMB trn % 30% 20% 10% 0% 62% 55% 41% 47% 27% 14% H onwards Source: Deutsche Bank, CEIC, PBOC Source: Deutsche Bank, NAO The refinancing of the Rmb2trn of LGFV loans maturing in 2011 and 2012 (Figure 18) has also played a part in the growth of the shadow banking system, which is in part being fueled and funded by joint-stock banks burdened by the over-regulation of China s banking sector and the maintenance of a 75% LDR cap. We explore this more under the heading Relationship between banks and shadow banking system. Trust companies are playing a key role in growth Over the past three quarters, trust loans have contributed over 10% of the new social financing in China (vs. average of 4.9%), indicating their importance in the growth and spread of the shadow banking system in China. While regulated by the CBRC, trust companies have been growing in financial importance over the past three years, and have seen AUM jump to Rmb9.5trn in 1H13 from Rmb2trn in Figure 19: Trust loans as a percentage of total social financing 16% 14% Trust loan as % of total social financing Average 14.6% 13.4% 12% 10% Average (2006-2Q13): 4.9% 9.2% 10.2% 8% 6% 4.5% 4.6% 4.1% 4% 2% 1.9% 2.9% 3.1% 2.8% 1.6% 0% Q12 2Q12 3Q12 4Q12 1Q13 2Q13 Source: Deutsche Bank, CEIC Figures show that the net profits of trust companies posted a CAGR of 38% between 2009 and 2012, to reach Rmb35.5bn as of December 2012, while equity saw a CAGR of 23% during the same period. Trust companies, which are usually backed by strong parents like CITIC, CCB, and Ping An, had Page 11

12 Chinese Stay Positive total assets of Rmb252bn as of June 2013 (three-year CAGR of 21%), but managed assets worth Rmb9.5trn. Figure 20: Net profit of trust companies in China Net profit (LHS) YoY (RHS) RMB bn % CAGR: 37.7% % % 11% % 50% 45% 40% 35% 30% 25% 20% 15% 10% Figure 21: Total equity vs. ROE Total equity of trust companies (LHS) ROE (RHS) RMB bn % 20% H13 CAGR: 23.3% % % 16.5% % % % % 14% 12.6% % % % % H13 Figure 22: Total assets vs. AUM Total assets of trust companies (LHS) AUM (RHS) RMB bn RMB trn CAGR: 20.8% H13 Source: Deutsche Bank, China Trustee Association, KPMG Source: Deutsche Bank, China Trustee Association, KPMG Source: Deutsche Bank, China Trustee Association, KPMG To understand the importance of trust companies to the growth of non-bank financing sector in China, we analyze the growth of the AUM of trust companies versus other non-bank financing sectors (insurers, brokers and funds). As Figures 23 and 24 show, the AUM of trust companies, which have risen close to 5x since 2009, are the highest in the non-bank financing sector (Rmb7.5trn for FY12). This compares with the AUM of Rmb7.36trn for Chinese insurers and Rmb1.9trn for Chinese brokers as of Dec Figure 23: Growth of AUM by Chinese financial companies ( ) RMB tn China Trust China Insurers China Brokers China Funds Source: Deutsche Bank, Asset Management Association of China, CBRC, CIRC, Securities Association of China, China Trustee Association Figure 24: AUM of trust companies vs. bank WMPs, insurers, mutual funds and brokers (2011 and 2012) RMB trn Bank WMPs Trust companies Insurers Mutual fund Brokers Source: Deutsche Bank, Asset Management Association of China, CBRC, CIRC, Securities Association of China, China Trustee Association 1.89 As Figure 25 shows, assets managed by trust funds are primarily divided into single fund trusts and collective fund trusts. As of June 2013, the outstanding assets managed by single fund trusts amounted to Rmb6.7trn or 71% of the total trust assets. Another Rmb2.1trn or 23% were invested under collective fund trusts. The risk weighting imposed on assets managed under collective fund trusts is generally higher than that for single fund trusts to reflect the more diversified ownership of risks amongst smaller investors. Approximately 69% of the assets managed under single fund trusts are made up of loans, followed by another 15% invested in available for sale & held to maturity investments (Figure 26). Page 12

13 Chinese Stay Positive Figure 25: Assets under management of trust companies breakdown by business RMB trn Collective Fund Trust: compound growth rate of 16.6%qoq Single Fund Trust: compound growth rate of 10.1% qoq Managing Property Trust: compound growth rate 9.7%qoq Figure 26: Breakdown of new investments of single fund trusts by product 100% Others % % % % 41 0% Due from other financial institutions Long Term Equity Investment AFS & HTM Investment Transactional Financial Asset Loan Source: Deutsche Bank, CEIC. Source: Deutsche Bank, CEIC, PBOC, CIRC, SAC, AMAC Figure 27 shows that investments for single funded trusts are primarily packaged loans lending to industry (29%), manufacturing (27%) and real estate (9%). Trust funds investments into the risky real estate sector have fallen from a high of 17% in 3Q11 to 9% in 2Q13. Figure 27: Breakdown of trust investment by industry 100% 80% 60% 40% 20% 0% Others Industrial & Commercial Enterprise Financial Institution Securities Market Real Estate Foundational Industry Figure 28: Trust funds investment in real estate RMB bn Trust fund invested to real estate (LHS) As % of total fund trust (RHS) % % 17% % 16% 18% % 13% % 11% 15% 14% % 13% 12% % % 10% % 9% 8% % 200 4% % 0% Source: Deutsche Bank, CEIC. Note: Foundation industry includes agriculture, manufacturing, light and heavy industries, infrastructure and services Source: Deutsche Bank, CEIC These trust companies usually function on a cooperation model, with trust investments guaranteed by the other entity be it a bank, local government or private entity. From 3Q10, government-trust cooperation has seen a CAGR of 7%, leading to strict monitoring and regulation of this relationship by the PBOC from 2012 onwards. Bank-trust cooperation, on the other hand, posted a CAGR of 1% over this period, and accounted for Rmb2.1trn as of June 2013, over one-fifth of the total trust cooperation, followed by Rmb0.80trn and Rmb0.26trn with the local government and private funds, respectively. In Appendix A, we mention the key policies and regulations that have affected the trust business since Page 13

14 Chinese Stay Positive Figure 29: Cooperation between trusts and other entities (3Q10 2Q13) RMB trn Bank-trust cooperation; CAGR: 0.9% qoq Trust-government cooperation; CAGR: 7% qoq Private fund cooperation; CAGR: 9.1% qoq Q10 4Q10 1Q11 2Q11 3Q11 4Q11 1Q12 2Q12 3Q12 4Q12 1Q13 2Q13 Source: Deutsche Bank, WIND, CEIC Figure 30: Cooperation between trusts and other entities (2Q13) Total cooperation between trust and other entities (2Q13): Rmb 9.5trn Source: Deutsche Bank, WIND, CEIC Others, 6.31tr, 67% Bank-Trust cooperation, 2.08tr, 22% Govt.-Trust cooperation, 0.80tr, 8% Private fund cooperation, 0.26tr, 3% Relationship between banks and shadow banking system The recent SHIBOR squeeze displayed the uneven distribution of liquidity in China s financial system, where the largest banks dominate deposit-gathering, while the smaller banks have been crowded out to look for other means of maintaining their liquidity. This has led to a system where the smaller jointstock Chinese banks are financing the shadow banking system through two main channels inter-bank lending and wealth management products (WMPs). cooperation with trust companies, which totaled Rmb2.1trn in 2Q13, is carried out by proprietary investments or selling third-party-issued wealth management products, which are retail deposit substitutes issued by banks asset management arms. As of June 2013, the outstanding amount of WMPs issued by banks reached Rmb9.08trn, up from Rmb1.7trn in 2009, representing a fivefold increase. Out of this, Rmb2.78trn was categorized as non-standardized products, which are invested in non-tradable investment assets that are managed by the shadow banking system, such as trust loans, account receivables financing, bank acceptance and alternative credit. Figure 31: Outstanding balance of WMPs issued by Chinese banks from 2009 to June 2013 RMB trn H13 Source: Deutsche Bank, CEIC Figure 32: Composition of new corporate non-bank financing RMB bn Loans (entrusted & trust) Bank Acceptance Bill Bonds Others 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0 Loans (entrusted and trust) up 130%yoy between 1H12 and 1H ,948 1,882 3,390 5,589 4,786 6,643 4,495 FY02 FY03 FY04 FY05 FY06 FY07 FY08 FY09 FY10 FY11 FY12 1H13 Source: Deutsche Bank, CEIC, PBOC Page 14

15 Chinese Stay Positive In Figure 33, we provide a breakdown of non-standardized WMPs by H-share banks, totaling Rmb1.7trn. Given the CBRC s new rules restricting the unchecked growth of non-standardized WMPs, only CNCB and MSB have no remaining capacity to issue more of these products, while the other banks have the collective capacity to issue WMPs worth Rmb224bn. Figure 33: Exposure of H-share banks to non-standardized WMPs as of 1H13 RMB bn ICBC CCB ABC BOC BoCom CMB CNCB MSB CRCB Total Non-standard credit WMP ,721 WMP balance 1,200 1, ,347 Total asset - 1H13 18,723 14,859 14,223 13,256 5,718 3,811 3,437 3, ,925 Non-standard credit WMP as % of total WMP 34.0% 30% 32% 32% 18% 35% 49% c.40% 10% 32% as % of total asset 2.18% 2.05% 1.61% 1.45% 2.16% 3.64% 5.88% 3.52% 0.57% 2.21% Remaining capacity of credit-backed WMPs 35% WMPs % of assets ,396 Source: Deutsche Bank, company data Figures 34 and 35 provide a breakdown of WMPs by type, with the total WMPs issued by H-share-listed banks reaching Rmb5.3trn or 6.9% of total assets and 114% of shareholders equity as of June As a result of the deposit-gathering dominance of the big four banks, the joint-stock banks need to rely on issuance of WMPs as alternatives for their funding requirements. This explains why joint-stock banks like MSB, CMB and CNCB have the highest exposure to off-balance sheet WMPs, with the exposure exceeding over 10% of CMB and CNCB s liabilities. Figure 34: H-share listed banks: Non-standardized and standardized WMPs as % of total assets (as of June 2013) % Non-standardised WMPs as % of assets Standardised WMPs as % of assets CNCB BoCom CMB MSB CCB ICBC CRCB ABC BOC H-Share Source: Deutsche Bank, company data Figure 35: Off-balance sheet WMPs as a percentage of total equity (as of June 2013) Non-standardized WMPs as % of equity Standardized WMPs as % of equity % CNCB CMB BoCom MSB ICBC CCB BOC ABC CRCB H share Source: Deutsche Bank, company data also provide inter-bank funds to the non-bank or shadow banking system through resale agreements between financial institutions, and these financial assets amounted to Rmb5.2trn for the listed banks as of 2012, of which Rmb1.25trn and Rmb122.1bn were backed by bills and trust beneficial rights, respectively. The corresponding exposure was Rmb4.58trn as of December 2012 and Rmb2.97trn as of December Page 15

16 Chinese Stay Positive Figure 36: Reverse repo arrangements for listed banks RMB bn 6,000 5,000 4,000 3,000 2,000 1, , ,617 2,888 2, ,733 1,016 1,122 4, , ,281 2,873 1,493 1,614 1, H13 Source: Deutsche Bank, company data Others Lease payment receivables Trust Beneficial Rights Loans Bills Debt Securities Figure 37: Composition of reverse repo arrangements for listed banks 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0% 0% 5% 4% 32% 63% 60% 35% 10% 2% 38% 50% 14% 17% 0% 0% 50% 35% 55% 27% H13 Source: Deutsche Bank, company data Others Lease payment receivables Trust Beneficial Rights Loans Bills Debt Securities Joint-stock banks are more vulnerable to liquidity and credit risks of the shadow banking system According to data reported by companies, we believe that joint-stock banks have been actively serving as intermediaries to obtain funds from the stateowned/city commercial banks and lending to other banks and non-bank financial institutions to make money on income spreads. In other words, the banking system has become increasingly exposed to the shadow banking system, and the credit risks of the latter might evolve into counterparty risks for the official banking system. For example, the outstanding value of Minsheng Bank s reverse repurchase agreement was as much as 57% of its customer base as of June Compare that to the big four banks, which saw a range between 1% and 11%. Figure 38: Interbank exposure of Chinese banks (1H13) 1H13 (RMB mn) ICBC CCB ABC BOC BoCom CMB CNCB MSB CRCB H share CEB Interbank Assets Deposits with banks and 303, , , , ,822 49, , ,161 21,045 2,732,786 73,870 non-bank fins Placements with banks and 338, , , , ,549 96,947 90,051 96,024 41,264 1,941, ,060 non-bank fins Financial assets held under resale agreements 462, , , , , , , ,719 53,704 3,424, ,384 Reverse repurchase analyzed by collateral: Debt Securities 340, , ,589 97,900 72, ,902 67,451 4,887 1,260 1,282,191 41,597 Bills 87, , ,263 12, , , , ,292 12,505 1,794, ,339 Loans 4, , ,007 Trust Beneficial Rights 103, ,540 39, ,290 Lease payment receivables 0 Others 29,222 8,241 19,480 56,943 2,448 Total reverse repo (net) 462, , , , , , , ,719 53,704 3,424, ,384 Total inter-bank asset 1,103,073 1,200,841 1,791,060 1,173, , , ,371 1,064, ,013 8,099, ,314 Total loans 9,437,642 8,095,052 6,946,120 7,439,633 3,201,417 2,052,599 1,824,552 1,484, ,329 40,674,314 1,104,554 Reverse repo % of total loan 5% 4% 11% 1% 6% 21% 12% 57% 28% 8% 16% Source: Deutsche Bank, company data Page 16

17 Chinese Stay Positive Our proprietary research trips in China suggest risks are high; but risk management seems effective amid stable economy Our proprietary research trips in China this July, when we met government officials and over 30 non-bank credit operators in Beijing, Shanghai, Hangzhou, Xi an, Shenyang, Hohhot, and Ordos, suggested that although the magnitude of the default risks is much higher for the less regulated non-bank financial system, the management of these risks seems entirely market-driven and generally effective. We believe the tighter regulation and de-risking of the official banking system have resulted in the emergence of the non-bank financial system to finance the assets that are perceived to be risky by the CBRC/shareholders (for example, LGFV, real estates, micro companies with little financial track record/cash flow issues). As such, the shadow banking system is inherently more risky, given the weaker quality of its assets, its limited ability to absorb losses due to the thinner capital base, its limited access to a stable source of funds and the more fragmented nature of the market. Similar to the official banking system, the key risks faced by the shadow banking system include liquidity risks, interest rate risks, credit risks and capital risks. However, given the key differences in the sourcing of funds and lending and risk management practices, these risks manifest differently in the shadow banking industry. The offsetting factors are (1) high lending rates charged by these companies to the borrowers to offset risks, (2) back-up by shareholders, and (3) policy intervention to avoid systematic risks. Liquidity risks Participants in the shadow banking system, especially the trust companies, secure funds through WMPs sold by banks/other FIs and by accessing the inter-bank markets, with the joint-stock banks serving as a conduit. As such, the system is vulnerable to risks arising from investors not rolling over the WMPs and regulations restricting growth and issuance of WMPs and the investments of the proceeds. These companies face additional risks if commercial banks scale back their inter-bank lending to other financial institutions. We believe liquidity risk represents the key risk to the shadow banking system, which might not be able to secure funds by paying up when risk aversion towards them rises. The liquidity risks will then magnify the default risks. Interest rate risks As loans tend to be short tenor and charged at a high fixed rate (for example, approximately 20% is charged by private money lenders at Wenzhou), interest rate risk is not the key risk faced by the system. This can be concluded by looking at the minimal impact on the average lending rate despite the SHIBOR spike in June. As Figure 39 shows, the average rate for private lending in Wenzhou for June (20.18%) was fractionally less than the reported level in the three months preceding it (20.39%). Moreover, the volatility in the daily rate was not as pronounced as had been feared, showing perhaps that the lenders are already charging a premium. Page 17

18 Chinese Stay Positive Figure 39: Wenzhou one-month private lending rate % Figure 40: Wenzhou private lending composite rate % Source: Deutsche Bank, PBOC, CEIC, WIND Source: Deutsche Bank, WIND, CEIC, PBOC Note: Date for Jan 11-Sept 11 is from news reports, 4Q11-3Q12 is from WIND, 4Q12 from PBOC, 1Q13-2Q13 calculated as simple daily average Credit/default risk We believe the credit/default risks are inherently higher for the non-bank system, as their borrowers tend to be more risky, making the system more vulnerable to an economic downturn. However, we believe the credit risks might be better micro-managed, as the system is entirely commercially driven with the market determining lending rates. Given the limited capital and customer reach, there is a stronger profit incentive for the small money lenders to assess the risks of borrowers in order to avoid capital losses. Our proprietary research trips in China (Shanghai, Hangzhou, Xi an, Shenyang and Beijing), which covered more than 30 non-bank credit operators, also suggest that although the magnitude of the default risks might be higher, the management of risks seems effective. Capital risk Amongst the different operations in the non-bank system, we believe the capital risk is the highest for the trust companies, as they set aside only limited capital for the assets under their management (a leverage ratio of 36x AUM/equity) as the intermediary between investors and borrowers. As such, the trust companies might not have sufficient capital to absorb losses if they are asked to by regulators. In the past, there have been precedents when the trust companies were asked to bail out sour investments. As a result, there is little clarity over the ownership of risks in the event of default for trust products. Page 18

19 Chinese Stay Positive Figure 41: Adjusted leverage ratio of trust sector (AUM/equity) (x) Source: Deutsche Bank, China Trustee Association, KPMG Figure 42: Total equity of trust companies vs. ROE RMB bn Total equity of trust companies (LHS) ROE (RHS) % H13 CAGR: 23.3% % % 12.6% H13 Source: Deutsche Bank, China Trustee Association, KPMG 20% 19% 18% 17% 16% 15% 14% 13% 12% 11% 10% Shadow banking is transforming the risk profile of the banking system CBRC could effectively slow the flow While tighter bank regulations have led to lower credit and market risks for the listed banks, it has also contributed to uncharted growth in the shadow banking system from Rmb5.8trn in 2010 to Rmb12.5trn as of June As a result of banks reducing exposure to risky sectors (LGFV, real estates and others) and shortening loan duration to match liabilities, the shadow banking system is absorbing the supply imbalance of corporate loans, in addition to its original function of lending to small and micro companies. The systematic concerns raised by the growth of the shadow banking system arise from the reliance of shadow banks on funding from the banking system, as this high connectivity means credit or liquidity risks in shadow banking could evolve into counterparty risks for the banks. Amongst the listed banks, the largest commercial banks (about 45% of banking system assets) have proportionately the lowest exposure to shadow banking. The three policy banks, which account for about 8% of assets, are pure sovereign risk. The risks appear to be greatest among the joint-stock banks (18% of assets), which are the main conduits/intermediaries between the banking system and the non-banks. They have disproportionately large exposures to WMPs and to inter-bank assets collateralized against bills and trust beneficial rights. It was this group of banks that was at greatest risk during the SHIBOR shock in June as wholesale funding dried up. We believe the CBRC could effectively slow the flow to the shadow banking system by increasing the risk weightings on resale agreements. Our analysis shows that if the weighting on resale agreements collateralized against bills was raised to 75% and that of trust beneficial rights to 100%, the CT1 of the H- share listed banks would be lowered by 23bps to 9.44%, with the biggest negative impact felt by MSB (-1.32% to 6.54%) and CRCB (-1.42% to 9.5%). Page 19

20 Chinese Stay Positive Figure 43: Change in CT1 ratio of banks if risk weighting of reverse repo backed by bills and trust beneficial rights rises to 75% and 100% respectively % bps Figure 44: New CT1 ratio of banks if risk weighting of reverse repo backed by bills and trust beneficial rights rises to 75% and 100% respectively Source: Deutsche Bank estimates, company data Source: Deutsche Bank estimates, company data Page 20

21 Chinese Stay Positive Valuation and risks Valuation of Chinese banks We value Chinese banks based on an ex-growth valuation (2018 onwards), computed using a three-stage Gordon Growth Model (PV= (ROE-g)/(COE-g)). Our ex-growth prices assume zero growth in book value from 2018 in perpetuity, and our target price calculations are based on 2013E book values. On our estimates, H-share-listed Chinese banks are trading at a 2013E P/B of 1.1x and 2013E P/E of 5.8x, offering 35% and 27% potential upside, respectively, assuming the sector trends back to its average historical valuation of 1.4x P/B and 7.4x P/E ( ). Our top picks are the big four Chinese banks, and the key downside risk to our view is that the banking sector s improving liquidity, driven by capital inflows, proves to be a necessary but insufficient condition for China s economy to recover. Our valuation assumes a near-term ( E) ROE of 13-19%, medium-term ( E) ROE of 12-16%, and terminal ROE of 9-12%, with COE of 11-12%. Figure 45 highlights our valuation comparisons for listed banks. Figure 45: Chinese banks valuation summary Ticker Rating TP Price Upside Mkt. Cap P/E (x) P/B (x) P/PPOP ROAE ROAA Div. Yield (%) (%) (US$mn) 12A 13E 14E 13E 14E 13E 14E 13E 14E 13E 14E 13E 14E ICBC-H 1398.HK Buy % 232, % 17.1% 1.35% 1.14% 4.92% 4.61% CCB-H 0939.HK Buy % 129, % 16.1% 1.39% 1.14% 5.06% 4.57% ABC-H 1288.HK Buy % 197, % 15.1% 1.18% 0.94% 4.98% 4.37% BOC-H 3988.HK Buy % 141, % 14.4% 0.99% 0.92% 5.28% 5.05% BCOM-H 3328.HK Buy % 55, % 13.8% 1.08% 0.99% 5.27% 5.32% CMB-H 3968.HK Buy % 47, % 15.6% 1.36% 1.12% 4.23% 3.88% CITIC Bank-H 0998.HK Buy % 29, % 15.6% 1.04% 1.07% 4.75% 5.44% Minsheng-H 1988.HK Hold % 44, % 19.2% 1.20% 1.07% 4.32% 4.24% CRCB 3618.HK Buy % 1, % 13.7% 1.25% 0.97% 6.06% 5.34% H-share sector mean % 15.8% 1.23% 1.04% 4.96% 4.63% Note: Closing price of Sep 16, 2013 Source: Deutsche Bank estimates, Bloomberg Finance LP Page 21